Documente Academic
Documente Profesional
Documente Cultură
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
a r t i c l e
i n f o
Article history:
Received 8 August 2014
Received in revised form 17 December 2014
Accepted 20 December 2014
Available online 29 December 2014
Keywords:
Methyl mercury
Advanced mercury analyzer
Marine biota
Sample preparation
Method validation
Traceability
Uncertainty
a b s t r a c t
In this paper, we present a simple, fast and cost-effective method for determination of methyl mercury
(MeHg) in marine samples. All important parameters inuencing the sample preparation process were
investigated and optimized. Full validation of the method was performed in accordance to the ISO17025 (ISO/IEC, 2005) and Eurachem guidelines. Blanks, selectivity, working range (0.093.0 ng),
recovery (92108%), intermediate precision (1.74.5%), traceability, limit of detection (0.009 ng), limit
of quantication (0.045 ng) and expanded uncertainty (15%, k = 2) were assessed. Estimation of the
uncertainty contribution of each parameter and the demonstration of traceability of measurement results
was provided as well. Furthermore, the selectivity of the method was studied by analyzing the same
sample extracts by advanced mercury analyzer (AMA) and gas chromatographyatomic uorescence
spectrometry (GCAFS).
Additional validation of the proposed procedure was effectuated by participation in the IAEA-461
worldwide inter-laboratory comparison exercises.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Mercury (Hg) is a persistent global pollutant, particularly in the
form of methylmercury (MeHg), a potent neurotoxin produced in
the aquatic environment from inorganic mercury by sulfate and
iron-reducing bacteria, as well as methanogens (Siciliano,
ODriscoll, & Lean, 2002). Indeed, MeHg is recognized as a major
environmental pollution issue and health hazard for humans
(Qiu, 2013). Owing to its capability to permeate through biological
membranes, once MeHg enters the food chain, it is efciently accumulated and transferred to organisms at higher trophic levels
(Mason & Benoit, 2003). As a result, the fraction of MeHg from
the total Hg (THg) in muscle tissue of top predator sh can be up
to almost 100% (Senn et al., 2010). The majority of population is
exposed to Hg through consumption of marine and freshwater
biota, mainly sh and seafood products (Sunderland, 2007). Specifically, large predatory sh which are at the top of the foodchain,
such as swordsh and tuna, contain high levels of MeHg and are
signicant sources of human exposure to that contaminant. Public
health warnings and guidelines on consumption of sh containing
high levels of MeHg have been published by the U.S. Food and Drug
Administration (USFDA, 2004) and the European Union (European
Corresponding author.
E-mail address: e.vasileva-veleva@iaea.org (E. Vassileva).
http://dx.doi.org/10.1016/j.foodchem.2014.12.085
0308-8146/ 2014 Elsevier Ltd. All rights reserved.
368
2. Experimental part
2.1. Instrumentation
2.3.1. Procedure 1
A 0.10.8 g portion of sample (or CRM) was weighted in a 50 mL
polypropylene centrifuge tube; 5 mL of 25% (v/v) HCl solution were
added and the mixture was vigorously shaken for 30 s. Then, 10 mL
of toluene were added in the tube and a vortex shaking method
was subsequently employed for 3 min to ensure the homogenization of the phases. The mixture was centrifuged at 5000 rpm for
520 min. A known volume (48 mL) of the upper organic phase
was removed and transferred to a second 50 mL polypropylene
centrifuge tube containing 510 mL of 0.002 M sodium thiosulfate
solution. This second tube was vigorously shaken (vortex, 3 min)
and centrifuged at 5000 rpm for 515 min. Two milliliters of the
lower aqueous phase, which contains the extracted organic mercury, was transferred to a 15 mL polypropylene container or glass
vial, using a Pasteur glass pipette. Then, an aliquot (50400 lL)
of the extract was directly analyzed with AMA. The thiosulfate
extract was found to be stable for 2 days at temperature of 4 C.
2.3.2. Procedure 2
0.10.5 g of sample was placed into a 50 mL polypropylene centrifuge tube and hydrolyzed with 10 mL of HBr. Twenty milliliter of
toluene was added and the mixture was homogenized for 2 min
and centrifuged for 10 min at 3000 rpm. The organic phase was
transferred to a tube containing 6.0 mL of 1% L-cysteine solution.
A second organic extraction was subsequently performed. A
0.5 mL aliquot of L-cysteine extract was immediately analyzed with
the AMA.
This procedure was proposed by the European Commission as a
standard operation procedure (SOP) for determination of MeHg by
direct mercury analyzer in sea food to all European Reference Laboratories for trace elements in food and feed (Caldern et al., 2013).
2.4. Determination of moisture content
Correction for dry-mass was obtained from 3 sub samples of
biota sample of minimum mass of 1.0 g. The material was dried
for 24 h in a ventilated oven at a temperature of 85 2 C. Then
weighing and repeated drying was performed until constant mass
was attained (0.0002 g difference between two successive weighs).
The loss of mass corresponds to the dry mass correction factor,
which was applied for the estimation of the combined uncertainty.
2.5. Experimental set up for method validation
Several parameters were evaluated for the validation of the proposed procedure, namely: selectivity; trueness by recovery, repeatability and within-laboratory reproducibility, instrumental/
method detection limits (LoDs) and quantication limits (LoQs),
range of linearity, measurement uncertainty, traceability of measurement results. Recovery, repeatability and intermediate precision were evaluated at 4 levels of concentrations. Furthermore,
stability studies of the solution were carried out.
The validation experiments were performed on six different
days. Independent samples were prepared on each single day.
Some of the experiments were used in the estimation of different
parameters.
During the validation study, an internal quality control (IQC)
procedure was adopted the blank level and the drift of the instrument readings were systematically checked analyzing one standard solution after every 10 samples.
Procedural blank was prepared together with unknown samples
in order to account for cross contaminations during the validation
study. It underwent the same analytical procedure as for biota
samples without adding biota matrix.
The calibration curve was prepared with several different mass
quantities, covering a range of mercury from 0.1 to 3 ng, in a x
369
370
3 and 5 min extraction time, the shortest time was selected. Overall, the extraction conditions were xed at 25% (v/v) HCl and 3 min
of shaking time for solvent and back extraction steps.
Ultraviolet (UV) and visible (Vis) radiations lead to degradation
of MeHg (Li et al., 2010). In this respect, parallel extractions, under
and without UV exposition, were conducted. No signicant differences in the recovery of MeHg from marine biota samples were
observed.
Although some published protocols (Caldern et al., 2013;
Maggi et al., 2009; Scerbo & Barghigiani, 1998; Valega et al.,
2006) recommend double solvent extraction, the quantitative
recoveries obtained for IAEA-436 using proposed methodology
demonstrates that single extraction is sufcient.
In order to avoid long centrifugation time, sample size and volume of back extracted toluene were kept as small as possible, but
sufcient to obtain a measurable signal in the measurement step.
In the case of formation of undesirable emulsion at the liquid
liquid interface, the nal volume of thiosulfate solution was
increased. The latter had to be optimized as a function of the volume of measured sample aliquot, because measurement time with
AMA strongly depends from the volume of the sample used in the
measurement step. Measurement time was from 4 to almost 8 min
for aliquot volumes from 50 to 400 lL, respectively.
3.2. Stability study
The variation of analyte level in the thiosulfate extract may give
signicant biases of measurement result. The stability was rst
checked by repeated measurements of one sample extract kept
at room temperature during 5 h, 2 days after their preparation
(extracts are stored at 4 C in between). No statistical differences
were observed in the different determination and it was concluded
that extracted solutions are stable at room temperature for at least
5 h. Five hours is the usual duration of the measurement sequence.
Additionally solutions can be kept at 4 C up to 2 days before the
measurement.
3.3. Comparison of both extraction procedures
The extraction procedure proposed in this study was then compared with the extraction procedure described in procedure 2,
where MeHg was extracted following the standard operation procedure distributed for the collaborative study in EU (Caldern et al.,
2013). For this comparison three CRMs namely IAEA-436, IAEA-452
and TORT-2 covering a wide range of MeHg concentration, (i.e.
0.023.67 mg kg1) and MeHg/THg ratio (from 14% to 88%) in different biota matrices were analyzed by applying both extraction
procedures. Obtained results are presented in Table 1. MeHg mass
fraction yielded by the methodology developed in this study is in
very good agreement with those obtained with procedure 2.
One of the advantages of the proposed extraction protocol is the
signicant reduction of the generated waste volume of organic
solvent. In addition damages of sample boat and auto sampler
Table 1
Comparison of results obtained for MeHga content (X U, mg kg1)b in marine biota CRM with the extraction procedure developed in this study and different detection methods
with the results obtained with EC recommended method.
IAEA 436
DOLT-2
TORT-2
IAEA 452
a
Certied values
Proposed procedure
GCPyAFSc Procedure
EC recommended procedure
3.67 0.42
0.693 0.053
0.152 0.013
0.022 0.004
3.39 0.51
0.743 0.111
0.149 0.022
0.021 0.003
3.80 0.45
0.737 0.088
0.017 0.002
3.75 0.45
0.147 0.022
0.019 0.005
Reported as mercury.
Uncertainties are reported as expanded uncertainties U = kuc (k = 2). Expanded uncertainties of the method proposed by EC were taken from the report of the collaborative study.
c
As described in Carrasco and Vassileva (2014).
b
371
Fig. 1. Typical chromatograms (a) for the separation of mercury and MeHg from CRM IAEA-452 subjected to the optimized extraction procedure, (b) for the instrumental
blank and (c) for mercury and MeHg extracted by traditional alkaline digestion with 25% (w/v) KOH in methanol from CRM IAEA-452.
372
130%
120%
Recovery
110%
100%
90%
80%
70%
0.0
0.5
1.0
1.5
2.0
2.5
W(MeHg) in mg kg-1 as Hg
Fig. 2. Accuracy prole.
3.0
3.5
4.0
373
CD i CM
mi1
mM
m1
mM md 1 1 m1 md 2 2
mi1 md i i
Table 3
Single laboratory validation approach for the calculation of combined uncertainty of
MeHg mass fraction in biota sample.
Combined standard uncertainty (as relative)
uc
q
u2Rw u2bias
C meas
CD
i1
AS Aicorr ABlk C D i Ai1corr AS ABlk
Ai1corr Aicorr
2
uRw SRw
Absorbance correction
Acorr Astd AI
Blk
Recovery calculation
ubias
n
1 X
C CRM n
R
C CRM cert
n
1
Parameter
s
Pn
2
i1 biasi
clab i cref i
cref i
s
Pp
2
i1 uc ref i
uc
5
ref
Index
Parameter
Index
Stock solution
u
S
RMS
C
lab
n
Rw
ref
Absorbance
Recovery
CRM
V
W
Volume (mL)
Moisture content of
biota sample (%)
Dilution factor
Std
meas
Calibration standard
Measured
cert
n, p
Blk
I_Blk
Corr
Certied
Number of repeats
Procedural blank
Instrumental background
Correction for instrumental
background
Added toluene
Collected toluene
Volume of sodium thiosulfate
Measured aliquot thiosulfate
q
RMS2bias u2ref
biasi
t1
t2
thio
aliquot
Standard uncertainty
Standard deviation
Root mean Square
Mass fraction (mg kg1)
6
5
4
3
2
ILC
90
ID-ICP-MS
GC-Pyr-AFS
Org Hg AMA
80
70
60
50
40
30
ILC
Laboratory
Number of CRM used
Within laboratory reproducibility
Certied reference material
7
W(MeHg) mg kg-1 as Hg
RMSbias
W(MeHg) mg kg-1 as Hg
ID-ICP-MS
GC-Pyr-AFS
Org Hg AMA
Fig. 3. Comparison of the MeHg mass fractions in candidate CRM IAEA 470 (A) and
IAEA 461 (B) obtained from: the IAEA world-wide inter-laboratory comparison
(ILC), isotope-dilution inductively coupled plasma mass spectrometry (ID-ICP-MS);
gas chromatography coupled to atomic uorescence spectrometry (GCPyAFS)
and the method proposed in the present study (Org Hg AMA).
374
375